|Event: A change of state. The event itself has no duration, otherwise the beginning and the end of the event would have to have their own duration or the beginning and the end of an event in turn would be independent events. See also regress, process, flux, change, states._____________Annotation: The above characterizations of concepts are neither definitions nor exhausting presentations of problems related to them. Instead, they are intended to give a short introduction to the contributions below. – Lexicon of Arguments. |
Books on Amazon:
Event/Semantic Domain/Language Acquisition/Semantics/Gärdenfors: Thesis: I am modeling events with two vectors: a force vector, which typically represents an action, and a result vector that describes a change of a physical movement or an object.
Events/Gärdenfors: there are principally three different approaches:
(i) Metaphysical analyzes describing the ontology of events
(ii) Cognitive models that represent how humans (or animals) represent events mentally. See Langacker (1987, sec. 3.3), Givón (2001), Croft & Wood (2000), Langacker, (2008, chap. 3); Croft, (2012a, sec. 1.4).
(iii) Linguistic studies describing the expressions with which events are constructed.
Vectors/event/Gärdenfors: with vectors we can represent changes of objects and distinguish events from their linguistic expressions:
Definition State/Gärdenfors: is a set of points in a conceptual space.
Definition Change/Gärdenfors: a change of a state is represented by a vector.
Definition Path/Gärdenfors: is a continuous sequence of changes. (That is, there are no jumps).
Vectors: not all belong to the acting ones: e.g. opposing forces.
Acting/Agent: is not necessarily part of the event.
Gärdenfors: this is about mental representation, not about a scientific representation of what is happening in an event, e.g. physically.
Vectors: an event contains at least two vectors and one object. 1. Result vector: represents the change, 2. Force vector: causes the change.
Event/intransitive/Gärdenfors: Problem: in intransitive constructions (e.g. "Susanna goes") the acting and the changed object (patiens) are identical. Then the conceptual space of the agent and of the object (patiens) coincide.
Partial events/decomposition/parts/Gärdenfors: two ways can be selected when dividing into sub-events:
1. Events can be divided as simultaneously occurring or parallel partial events in the dimensions of the object space (patient space).
2. They can be represented successively by parts of paths.
Agent/Patient/semantic roles/Gärdenfors: both can be represented as points in the category space. The domains of the space then define the properties of both.
Patient/Linguistics/Gärdenfors: can be animated or inanimated, concrete or abstract. It has its own patient space with domains for properties. In contrast to the object categories, the properties usually contain the localization.
Agent: has accordingly its agent space, which has at least one force domain.
Dowty (1991): presents prototypical agents and prototypic patients. It is also about volitional involvement in an event.
Event/Linguistics/Gärdenfors: there are three approaches for dealing with events in linguistics:
1. Localist Approach: (Jackendoff, 1976, 1983, 1990): Thesis: all verbs can be constructed as verbs of movement and localization.
GärdenforsVsJackendoff: in his approach...
...force vectors cannot be represented appropriately.
2. Approach on aspects: (e.g. Vendler, 1957): distinguishes between states, activities, achievements and accomplishments. See also Jackendoff, 1991, sec. 8.3; Levin & Rappaport Hovav, 2005, p. 90).
3. Causal Approach: e.g. Croft (2012, a, 2012b) three-dimensional representation of causal and aspectual structures of events. Gärdenfors: that comes closest to my own approach. A geometric model is designed here.
The vectors in such models are not in a vacuum, but are always in relation to a domain and its information, e.g. temperature.
GärdenforsVsCroft: his approach does not support force vectors.
The Geometry of Meaning Cambridge 2014